Proof Of Concept For X-Ray Backscatter Imaging Tomosynthesis Using Spectral Detection

This paper describes a proof of concept for X-ray backscatter imaging beyond its common 2D radiograph capabilities by utilizing energy discrimination on detectors in an X-ray backscatter system. This study is a direct improvement over previous works, which required a separate scan for each depth being imaged; the current study achieves depth sensitive images with just a single scan. This technique is applicable to images generated from materials found in both medical and non-destructive testing environments. The system consists of a collimated pencil-beam of radiation surrounded by detectors that measure the fluence of backscattered photons coming from the object under investigation. Since X-ray tubes emit a spectra of radiation and the penetration depth of a photon is energy dependent, detectors using energy discrimination will measure signals generated from different depths within the object. The stochastic radiation transport code MCNP6 was used to simulate and test this concept. The simulated backscatter system used a 5 mm pencil-beam surrounded by four 5 cm diameter detectors, with the photon spectra modeled after a 150 kVp lead target X-ray tube. The system raster scanned over a phantom made of tissue that had the letters U and F embedded 6 cm and 8 cm deep, respectively. The letter U was made of air while the letter F was made of compact bone. The measured signals were discriminated into two energy groups, one consisting of photons less than 90 keV and the other anything above it. The two resulting images clearly proved the depth sensitive imaging capability, with the lower energies detected only showing the letter U, while the higher energies showed both the letters U and F. This technique should easily be applicable to industrial materials, though higher interrogation energies are likely needed.